Run 4. Outer Step 58. Inner Step 18.

config.json

@@ -268,18 +268,14 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5586689,
-    5586693,
-    5586697,
-    5586702,
-    5586707
+    5586740
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 17,
+  "inner_step": 18,
   "inner_steps": 0,
   "last_allreduce_block": 5584811,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c825c7af320df501fb5dee48cd40b9754a5d6e7773cc12e85a4cd8af86a4fb7
+oid sha256:b3d0352818ac71f239c681e3f26725017901745c162066ded171f9b9e640c7b0
 size 8081782503

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:174170927170b57e6919a3ed4d4904eb2c491cf919af568a6e6cbb72681b87be
+oid sha256:0fbdddddefbd3c1a34392f318c8f8a3c172a6f6cfcf319486c42ce29bcb84507
 size 4040701744